Spain's Multiverse Computing raises €189m to shrink AI model size by 95%

Multiverse Computing, a Spanish AI company, has secured €189 million in funding led by Bullhound Capital, HPE, Forgepoint Capital, and Toshiba. The funding will support Multiverse's technology, which compresses large language models by up to 95% and reduces computing costs by 80% without sacrificing performance. Multiverse has already released compressed versions of models like Meta’s Llama and Mistral, and the compression tool is available on Amazon Web Services.

Analysis

Spanish artificial intelligence firm Multiverse Computing has secured €189 million in a significant funding round led by Bullhound Capital, with participation from Hewlett Packard Enterprise (HPE), Forgepoint Capital, and Toshiba. This capital infusion is earmarked to advance Multiverse's proprietary technology, which aims to compress large language models (LLMs) by as much as 95% and reduce associated computing costs by up to 80%, all while reportedly maintaining performance integrity. The company's approach, which leverages principles from quantum physics and machine learning, notably operates on classical computing systems, enhancing its immediate applicability. Multiverse has already made tangible progress by releasing compressed iterations of widely-used open-source models, including Meta’s Llama, DeepSeek, and Mistral, and its compression tool is accessible through Amazon Web Services’ AI marketplace. This funding round positions Multiverse as Spain’s largest AI startup and a prominent European entity in the sector, with a strategic focus on optimizing models currently utilized by enterprise clients. The general sentiment surrounding this development is strongly positive (0.85), with a market impact score of 0.65, reflecting optimism about its potential to address key efficiency and cost barriers in AI deployment. The involvement of HPE and distribution via AWS, both of which register a mildly positive sentiment (0.5) from this news, adds credibility and potential go-to-market strength.

Key Decisions for Investors

• Investors should monitor the adoption and performance validation of Multiverse's LLM compression technology, particularly through channels like Amazon Web Services, as its success could materially reduce AI operational expenditures for enterprises and influence the competitive dynamics among AI service providers.
• The strategic investment by Hewlett Packard Enterprise (HPE) warrants attention from HPE investors, as it may signal future integration of Multiverse's technology into HPE's AI EaaS (Enterprise AI as a Service) offerings, potentially enhancing its competitive edge in the AI infrastructure market.
• Given the potential for significant cost reductions in AI model deployment, investors should consider the broader implications for companies heavily reliant on large-scale AI, as this technology could lower barriers to entry and accelerate AI adoption across various industries if widely successful.